System, method and program to test a web site

- IBM

Computer system, method and program product for testing a web site are disclosed. Numbers of hyperlinks required to navigate from one web page of the web site to other, respective web pages on the web site are determined. A score for ease of navigation though the web site based at least in part on the numbers is generated, such that lower numbers correlate to easier navigation through the web site. Another computer system, method and program product for testing a web site are disclosed. First keyword searches are conducted in the web site, using a search engine in the web site, based on sets of search terms ANDed together. Second keyword searches are conducted in the web site, using the search engine at the web site, based on respective subsets of the sets of search terms. Each subset is either a single term within the respective set or two or more terms within the respective set ANDed together. A score for effectiveness of the search engine is determined based at least in part on the frequency at which search results of the first keyword searches are contained in search results for the respective second keyword searches.

Skip to: Description  ·  Claims  · Patent History  ·  Patent History
Description
FIELD OF THE INVENTION

The invention relates generally to computer systems and the World Wide Web, and more specifically to a program tool to test a web site.

BACKGROUND OF THE INVENTION

The World Wide Web is well known today and comprises a vast multitude of web servers forming web sites. Client computers with web browsers are connected to the web servers via communication media, Internet Service Providers, firewalls, routers, etc. Upon request by a client computer, a web site can furnish a “home page” or other web pages to the client computer. The web browser on the client computer displays the home page or other web page for the user to view and enter data or make selections. Some of the selections are links to navigate through web pages in the same web site or other web sites.

Some complex web sites offer a local search engine to search through the web site based on key words entered by the user. The user can use the search engine to locate a web page of interest to the user. For example, the user may want to locate and display a web page containing information of interest to the user or a web page used to purchase a product. Some search engines are more effective than others, i.e. faster, more user friendly and better able to locate the web page of interest to the user.

Some web sites provide “support” to customers in their use of various hardware or software products. For example, Microsoft Corporation provides a web site to provide support or help to customers of various Microsoft operating systems. Likewise, IBM provides a web site to provide support or help to customers of various IBM program products such as Tivoli Storage Manager (“TSM”) web site and Lotus Notes web site. Many other software manufacturers also provide web sites for support or help to their customers.

An object of the present invention is to provide a program tool which automatically tests and determines usability of a web site.

Another object of the present invention is to provide a program tool which automatically tests and determines usability of a web site providing customer support or help for a manufacturer's hardware or software products.

SUMMARY OF THE INVENTION

The present invention resides in a computer system, method and program product for testing a web site. Numbers of hyperlinks required to navigate from one web page of the web site to other, respective web pages on the web site are determined. A score for ease of navigation though the web site based at least in part on the numbers is generated, such that lower numbers correlate to easier navigation through the web site.

According to features of the present invention, the one web page is a home page for the web site. The other web pages include questionnaires for a user to provide feedback about the other web pages, respectively.

The present invention also resides in another computer system, method and program product for testing a web site. First keyword searches are conducted in the web site, using a search engine in the web site, based on sets of search terms ANDed together. Second keyword searches are conducted in the web site, using the search engine at the web site, based on respective subsets of the sets of search terms. Each subset is either a single term within the respective set or two or more terms within the respective set ANDed together. A score for effectiveness of the search engine is determined based at least in part on the frequency at which search results of the first keyword searches are contained in search results for the respective second keyword searches.

According to features of the present invention, third keyword searches are conducted in the web site, using the search engine in the web site, based on predefined search terms suited for a type of the web site. A frequency at which the search engine identifies one or more hyperlinks complying with the third keyword searches is determined. A score for content of the web site based at least in part on the frequency at which the search engine identifies one or more hyperlinks complying with the third keyword searches is determined.

BRIEF DESCRIPTION OF THE FIGURES

FIG. 1 is a block diagram of a distributed computer system, including a web site test program tool which embodies the present invention.

FIG. 2 is a flow chart of the web site test program tool of FIG. 1.

FIG. 3 is a more detailed flow chart of part of the web site test program of FIG. 2 which determines a minimum number of intervening hyperlinks to navigate from a home page to each key web page or web file of the web site.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

The present invention will now be described in detail with reference to the figures. FIG. 1 illustrates a distributed computer system generally 10 comprising client computers 12a,b,c, web servers 14a,b,c, and a web site test server 20, all interconnected via Internet 21. Each of the client computers 12a,b,c includes known CPU 22a,b,c, operating system 24a,b,c, RAM 26a,b,c, ROM 28a,b,c, storage 30a,b,c and web browser 31a,b,c, respectively. Each of the web servers 14a,b,c includes known CPU 32a,b,c, operating system 34a,b,c, RAM 36a,b,c, ROM 38a,b,c and storage 40a,b,c, respectively. Each of the web servers 14a,b,c also includes web site hosting program 45a,b,c and database 42a,b,c of web pages and web files forming respective web sites. Each of the web pages is stored in HTML or other markup language, and each of the web files is stored in HTML or other format. Each of the web servers 14a,b,c also includes a respective counter program 47a,b,c to count the number of client requests to access each web page. By way of example, the web sites provide customer support or help for hardware and/or software products sold or licensed by the owners of the respective web sites. Many of the web pages in each web site provide a respective questionnaire (in pop-up or other format) to users of the respective web page, to elicit feedback from the users. (The questionnaires for the respective web pages also reside in databases 42a,b,c.) For example, after a customer uses a web page, a Java applet embedded in the web page, will create a pop-up questionnaire to ask the customer how usable was the web page, how easily was the web page found, and how effective was the web page, and for customer comments. Counter programs 47a,b,c also count the number of client responses to feedback questionnaires for each web page. Thus, for each web page, counter programs 47a,b,c count the number of client accesses and number of client feedbacks. Both counts represent a level of user interest in each web page.

Each of the web servers 14a,b,c also includes a respective keyword search engine 46a,b,c to search through the respective web sites for web pages and web files of interest to the user. Web site test server 20 also includes known CPU 52, operating system 54, RAM 56, ROM 58, storage 60 and web browser 61 (preferably enabled for Sun Microsystems Java program). Web site test server 20 also has known, high-speed Internet access such as DSL or cable, so its measurements of the web site are not affected by its own speed of access to the Internet 21 and web servers 14a,b,c. Web site test server 20 also includes a web site test program 70 according to the present invention. As explained in more detail below, program 70 automatically tests various characteristics of web sites such as content and organization of web pages and hyperlinks within the web pages and also presence and effectiveness of their local search engine 46a,b,c to determine the usability of the web site. More specifically, program 70 tests and determines how easily customers can navigate to web pages of interest, i.e. how many web pages must be traversed via respective hyperlinks to navigate from the home page to the web page of interest. Program 70 also tests and determines whether the local search engine of each web server 14a,b,c responds well to both one and two word searches, there is sufficient breadth to the searches and there is sufficient content of the searches. These are all indicia of the effectiveness of the local search engine.

FIG. 2 illustrates the operation and function of web site test program 70 in more detail. The following explanation of program 70 is for testing a single web site on a single server, but will be repeated for each other web site on the same or different server. By way of example, program 70 includes a test server portion 72 that executes in test server 20 and an agent program 74 which can be written in Sun Microsystems Java language and executes at the web server 14a,b,c under test. In step 100, a user initiates program 70 and specifies a URL of a home page of a web site to be tested. In response, program 72 sends its agent program 74 to the web site 14a,b,c under test. First, program 72 checks if the web site 14a,b,c has a knowledge base utilizing a site search (step 102). Next, program 72 conducts a number of predefined searches using the search engine at the web site 14a,b,c based on predefined search queries and checks and records the time it takes to receive a response (step 104). If there is no web page at the web site 14a,b,c that complies with the search query, this is noted as well. By way of example, the predefined search queries comprise search terms typically found on a web page for each type of web site. An administrator previously defined these search queries for each type of web site to be tested. At the start of the test, the user specified the type of web site currently under test, so program 72 knows which search queries to conduct. In step 105, program 70 generates a score for the test of step 104, where a response to all search queries (with at least one hyper link for a web page which complies with the search query) in the shortest period of time receives the highest score, and either a slow response or no response with a hyper link for a web page that complies with the search query results in a lower score.

Next, program 70, using web browser 61, initiates a series of predefined one-word and two-word, keyword searches of the web site (step 108). Preferably, each of the two-word searches includes one of the one-word search terms ANDed with another search term. For example, if some of the one-word search terms are database, volume, fixpack, then the corresponding two-word search terms can be database and restore, volume and dismount, and fixpack and interim, respectively. Program 72 identifies the key words for these searches from a file (previously defined by an administrator) where the one-word and two-word search terms are classified by type of web site. As noted above, when initiating program 70, the user specified the type of web site. Program 72 then compares the results (i.e. web page hyperlinks) of the one-word versus two-word searches as follows (step 110). If the search engine is working properly, each “hit” or result (i.e. web page hyper link) of a two-word search should be a subset of the “hits” or results (i.e. web page hyper link) of the corresponding one-word search, because the two-word search includes the one-word and the words of the two-word search are ANDed together. So, program 70 determines if each of the top ten (or other predetermined number) of hits of the two-word search are found in the hits of the corresponding one-word search. Next, program 70 generates a score for this comparison based on the number of hits of the two-word search that are found in the hits of the one-word search (step 112). The higher the number of hits of the two-word search that are found in the hits of the one-word search the higher the score.

Next, program 70 measures the ease of navigation though the web site as follows. Agent program 74 fetches the home page for the web site and counts the number of hyperlinks on the home page (step 120). (The hyperlinks are clickable representations of URLs of other web pages or web files, and can easily be counted from the corresponding HTML.) This is one indicia of ease of navigation through the web site because the hyperlinks on the home page facilitate access to other web pages or web files on the web site. Next, program 70 generates a score based on the number of hyperlinks of the home page, the greater the number of hyperlinks on the home page generally the easier to navigate intelligently through the web site (step 122). Next, program 70 measures the ease in accessing, via the home page, the other web pages or web files on the web site (step 130). This measurement is based on the number “clicks” or hyperlinks required to navigate to other, key web pages or web files from the home page. In this regard, each web site can be viewed as hierarchical, with each hyper link and respective web page being a node in the hierarchy. The home page is the highest node in the hierarchy. So, for a user to “navigate” from the home page to another web page requires traversal through the intervening hyperlinks and respective web pages. Ideally, each web page in the web site should be reachable from the home page with a minimum number of intervening hyperlinks. Accordingly, program 70 generates an ease of access score based on the shortest number of intervening hyperlinks to navigate from the home page to the key web pages or web files on the web site (step 140). The key web pages or web files can be (a) the web pages or web files that are most frequently accessed by users, (b) all web pages or web files at the web site, (c) web pages or web files previously designated by an administrator as key, or (d) those web pages or web files which receive the greatest number of feedback responses, i.e. those web pages or web files for which the greatest numbers of users have responded to a corresponding feedback questionnaire (such as a pop-up questionnaire for the respective web page or web file).

FIG. 3 illustrates step 140 in more detail, where program 70 determines the minimum number of intervening hyperlinks to navigate from the home page to each of the key web pages or web files. The agent program 74 at the web server reads the HTML for the home page (i.e. the first or highest level node in the hierarchy), and identifies each of the hyperlinks on the home page and respective web pages and web files (i.e. the second level nodes in the hierarchy) on the home page (step 200). Then, for each of the second level web pages, the agent program at the web server reads the corresponding HTML, and identifies each of the hyperlinks and respective web pages and web files (i.e. the third level nodes in the hierarchy) (step 202). Then, for each of the third level web pages, the agent program at the web server reads the corresponding HTML, and identifies each of the hyperlinks and respective web pages and web files (i.e. the fourth level nodes in the hierarchy) (step 204). The agent program 74 continues the foregoing process until all of the key web pages or web files have been identified in the hierarchy (decision 205 and step 206). (In the case where the “key” web pages or web files are all the web pages or web files on the web site, agent program 74 determines when all web pages or web files have been identified in the hierarchy by unique URLs.) Next, agent program 74 determines the shortest node-to-node path to each key web page or web file at the web site (step 208). The shortest node-to-node path is the path with the fewest intervening hyperlinks between the home page and the key web page or web file. Next, the agent program 74 reports back to program 72 the shortest node-to-node paths in the hierarchy from the home page to each key web page or web file (step 210).

Referring again to FIG. 2, next, program 70 determines a composite score for the web site, based on the separate scores generated in steps 105, 112, 122 and 140 (step 150). The composite score can be the summation of the separate scores, with appropriate weight, such as the following.

A×(ease of searching score)+B×(degree of two-word subset score)+C×(number of hyper links on home page score)+D×(ease of navigation score)=web site effectiveness score, where A, B, C and D are weighting constants.

Program 70 can be loaded into test server 20 from a computer readable media 92 such as magnetic disk or tape, optical disk, DVD, memory stick, etc. or downloaded from the Internet via TCP/IP card 94.

Based on the foregoing, system, method and program for testing a web site have been disclosed. However, numerous modifications and substitutions can be made without deviating from the scope of the present invention. Therefore, the present invention has been disclosed by way of example and not limitation, and reference should be made to the following claims to determine the scope of the present invention.

Claims

1. A method for testing a web site, said method comprising the steps of:

determining numbers of hyperlinks required to navigate from one web page of the web site to other, respective web pages on said web site; and
generating a score for ease of navigation though said web site based at least in part on said numbers, such that lower numbers correlate to easier navigation through said web site.

2. A method as set forth in claim 1 wherein said one web page is a home page for said web site.

3. A method as set forth in claim 1 wherein said other web pages include questionnaires for a user to provide feedback about said other web pages, respectively.

4. A method as set forth in claim 3 further comprising the step of displaying said questionnaires in pop-up windows associated with said respective other web pages.

5. A method as set forth in claim 1 further comprising the step of determining a number of hyperlinks on said home page, and wherein the step of generating said score is also based in part on the number of hyperlinks on said home page.

6. A method as set forth in claim 1 further comprising the steps of:

conducting first keyword searches in said web site, using a search engine in said web site, based on sets of search terms ANDed together;
conducting second keyword searches in said web site, using said search engine at said web site, based on respective subsets of said sets of search terms, wherein each subset is either a single term within the respective set or two or more terms within the respective set ANDed together; and
determining a score for effectiveness of said search engine based at least in part on the frequency at which search results of said first keyword searches are contained in search results for the respective second keyword searches.

7. A method as set forth in claim 6 further comprising the steps of:

conducting third keyword searches in said web site, using said search engine in said web site, based on predefined search terms suited for a type of said web site;
determining a frequency at which said search engine identifies one or more hyperlinks complying with said third keyword searches; and
determining a score for content of said web site based at least in part on the frequency at which said search engine identifies one or more hyperlinks complying with said third keyword searches.

8. A method for testing a web site, said method comprising the steps of:

conducting first keyword searches in said web site, using a search engine in said web site, based on sets of search terms ANDed together;
conducting second keyword searches in said web site, using said search engine at said web site, based on respective subsets of said sets of search terms, wherein each subset is either a single term within the respective set or two or more terms within the respective set ANDed together; and
determining a score for effectiveness of said search engine based at least in part on the frequency at which search results of said first keyword searches are contained in search results for the respective second keyword searches.

9. A method as set forth in claim 8 further comprising the steps of:

conducting third keyword searches in said web site, using said search engine in said web site, based on predefined search terms suited for a type of said web site;
determining a frequency at which said search engine identifies one or more hyperlinks complying with said third keyword searches; and
determining a score for content of said web site based at least in part on the frequency at which said search engine identifies one or more hyperlinks complying with said third keyword searches.

10. A computer program product for testing a web site, said computer program product comprising:

a computer readable media;
first program instructions to determine numbers of hyperlinks required to navigate from one web page of the web site to other, respective web pages on said web site; and
second program instructions to generate a score for ease of navigation though said web site based at least in part on said numbers, such that lower numbers correlate to easier navigation through said web site; and wherein
said first and second program instructions are stored on said media in functional form.

11. A computer program product as set forth in claim 10 wherein said one web page is a home page for said web site.

12. A computer program product as set forth in claim 10 wherein said other web pages include questionnaires for a user to provide feedback about said other web pages, respectively.

13. A computer program product as set forth in claim 12 further comprising third program instructions to initiate display of said questionnaires in pop-up windows associated with said respective other web pages; and wherein said third program instructions are stored on said media in functional form.

14. A computer program product as set forth in claim 10 further comprising third program instructions to determine a number of hyperlinks on said home page, and wherein said second program instructions generate said score based in part on the number of hyperlinks on said home page; and wherein said third program instructions are stored on said media in functional form.

15. A computer program product as set forth in claim 10 further comprising:

third program instructions to conduct first keyword searches in said web site, using a search engine in said web site, based on sets of search terms ANDed together;
fourth program instructions to conduct second keyword searches in said web site, using said search engine at said web site, based on respective subsets of said sets of search terms, wherein each subset is either a single term within the respective set or two or more terms within the respective set ANDed together; and
fifth program instructions to determine a score for effectiveness of said search engine based at least in part on the frequency at which search results of said first keyword searches are contained in search results for the respective second keyword searches; and wherein
said third, fourth and fifth program instructions are stored on said media in functional form.

16. A computer program product as set forth in claim 15 further comprising:

sixth program instructions to conduct third keyword searches in said web site, using said search engine in said web site, based on predefined search terms suited for a type of said web site;
seventh program instructions to determine a frequency at which said search engine identifies one or more hyperlinks complying with said third keyword searches; and
eighth program instructions to determine a score for content of said web site based at least in part on the frequency at which said search engine identifies one or more hyperlinks complying with said third keyword searches; and wherein said sixth, seventh and eighth program instructions are stored on said media in functional form.
Patent History
Publication number: 20070239729
Type: Application
Filed: Mar 30, 2006
Publication Date: Oct 11, 2007
Patent Grant number: 8166027
Applicant: International Business Machines Corporation (Armonk, NY)
Inventor: Salvatore Giallanza (Marana, AZ)
Application Number: 11/393,976
Classifications
Current U.S. Class: 707/10.000
International Classification: G06F 17/30 (20060101);